The ever-developing industry of nanotechnology the last two decades has culminated in a plethora of new nano-objects (defined as material with one, two or three external dimensions in the nanoscale) which are being used within a variety of consumer and industrial applications. Among the most prominent nano-objects are carbon nanotubes (CNTs); hollow nanofibres formed from carbon
. Based on their structure CNTs are classified as either, single-wall carbon nanotubes (SWCNTs), which comprise a single layer of carbon atoms, or multi-wall carbon nanotubes (MWCNTs), comprising of multiple concentric tubes
. Structural and mechanical characteristics such as an extreme strength, stiffness and robustness
 make CNTs interesting for the use in an infinite number of applications such as sporting goods, automobile products or household items. Additionally, CNTs hold great promise for application within medicine, particularly as a tool in therapeutics and diagnostics
. With their increasing number of applications, CNT emissions into the environment and human exposure may increase. Mainly CNT production, processing and disposal may be hazardous for humans
. Moreover during the use of CNT containing products, CNTs may be released into the environment as for instance from abrasion or degradation of CNT containing products. Possible portals by which CNTs may enter the human body include the skin, the gastro-intestinal tract and injection (nanomedicine). However, it is well accepted from previous research using nano-sized particles
 and CNTs
[7, 8] that inhalation is the primary exposure route to the human body if CNTs are released into the environmental air.
Concerns about the safety of CNTs have been raised for a number of different reasons
[3, 9, 10] (i) due to their small aerodynamic diameter CNTs are hypothesised to reach the lower respiratory tract, (ii) CNTs possess, like other nano-objects, a high surface to mass ratio, thus a large surface can interact with the biological surroundings, and (iii) some CNTs which are fibre shaped may (if structured dimensions are similar) behave like asbestos, or other pathogenic fibres which are toxic due to their needle-like shape. Moreover (iv), numerous in vivo studies (e.g.)
[11–13] have shown different types of MWCNTs to remain in the lung for up to several months after deposition indicating the potential for prolonged biopersistence.
Recently the potential adverse effects of CNTs have been studied on various biological systems, using different exposure methods both in vivo and in vitro
[14, 15]. Despite the unrealistically high doses which have been used within some of the previous studies (e.g.)
, it is known that subpleural fibrosis
, granuloma formation
 and mesothelioma
 similar to the effects of crocidolite asbestos fibres, can appear after in vivo exposures to mainly straight, stiff and extremely long CNTs.
Observed adverse effects have further been explained by the oxidative stress paradigm
. In numerous studies (e.g.)
[19–21] an increased oxidative stress response in vivo and in vitro has been reported causing a subsequent (pro-) inflammatory reaction after exposures to both straight and tangled CNTs.
Although numerous studies address the adverse potential of CNTs, their comparability is often limited and results are contradictory. Explanations for these discrepancies include differences in administered dose, the physico-chemical characteristics (e.g. agglomeration/aggregation state, metal impurities, stiffness, length) of the CNTs studied, the exposure method of CNTs, or differences in the biological system employed
[15, 22, 23]. Thus conditions/characteristics have to be manipulated systematically in order to identify key factors for their potential (adverse) biological effects. A promising way to modify the properties of CNTs is the functionalization of the surface
[24, 25]. Functionalization of CNTs can be used to promote the binding of specific biomolecules (such as siRNA)
 but also to improve their biocompatibility. The adverse effect potential of CNTs can be significantly driven by the particular (surface) modification employed
 whereas studies have shown both, increases and decreases in toxicity after exposures to CNTs with different surface functionalizations
[27, 28]. Functionalization can further affect the CNTs dispersity which can have subsequent consequences on their cell uptake and agglomeration in tissue
It is not only artificial surface modifications that play a role in regards to the potential adverse effects of CNTs. The surface characteristics of MWCNTs may also be modified by the adsorption of biomolecules following inhalation, and the subsequent interaction with the lung. Specifically, an initial coating of the MWCNTs will take place when they interact with pulmonary surfactant which is mainly produced by epithelial type II cells and which is located at the air-liquid interface. Surfactant consists 85-90% of phospholipids
, the specific surfactant proteins (SP) -A, -B, -C, and -D (~10%) and its main function is the reduction of the alveolar surface tension and keeping the gas exchange surface at optimal size during the movements of breathing
. Thus, during deposition, surfactant or surfactant components will bind to the surface of MWCNTs
[32, 33]. Previously, this initial coating has not been sufficiently considered in respect to in vitro lung toxicity studies. The modulation of the adverse potential from surfactant binding is mainly described for microparticles
, however to the best of our knowledge, not for CNTs and other nano-objects. After inhaled CNTs are coated with pulmonary surfactant, they may be displaced into the aqueous hypophase
[35–37] and come in contact with cells of the immune system such as macrophages and dendritic cells, which may engulf the MWCNTs and clear them from this area of the lung
. Also epithelial type I cells can interact with CNTs, as these cells mainly cover the alveolar surface
Objectives of the study and methodological approach
The primary aim of this study therefore, was to investigate how a pre-coating of MWCNTs with pulmonary surfactant may affect their potential adverse effects on cells of the air blood tissue barrier in vitro. In order to simulate the pulmonary surfactant coating, MWCNTs were pre-coated with Curosurf, a well characterized natural porcine surfactant preparation
Monocyte derived macrophage (MDM) monocultures as well as a sophisticated 3D in vitro triple cell co-culture model of the airway epithelial barrier
 were exposed to the MWCNTs either pre-coated with surfactant or not. MWCNTs were initially compared to their potential to affect cell viability. Subsequently, in order to study an early oxidative reaction, reactive oxygen species (ROS) was quantified in MDM. For the characterization of a prolonged oxidative stress response the intracellular antioxidant glutathione was quantified. The release of the cytokine TNF-α, as well as the chemokine IL-8 was quantified in order to study their possible (pro-) inflammatory effects.
The secondary aim was to investigate the influence of the surface charge on the MWCNTs potential adverse effects. Therefore, non-functionalized (“pristine” or “P-MWCNTs”), carboxyl (“MWCNT-COOH”) and amino (“MWCNT-NH2”) functionalized MWCNTs were used in the different exposures. In order to study a concentration dependence of potential adverse effects, 2–3 different concentrations (depending on the endpoint, see methods section for details) were applied. Eventually the different conditions (Curosurf pre-coating, functionalization, concentration) were statistically compared.